View Full Version : New speaker issue


Renton Maclachlan
June 8th, 2015, 01:50 AM
I took the plunge and got some Wharfedale 8.1 Studio Monitors so that I could get some better sound when editing a video project than I have been getting from computer speakers...

They turned up this morning and a friend lent me cabling for them so I could get them going. The cable has a mini jack to go into the computer (actually I have a mini-jack cable coming from the computer, into which I plug either speakers - past or present ones, or my headphones for when the house needs to be quiet), which switches to two XLR plugs to go into the balanced inputs in the new speakers.

However there is a teething issue in getting them going correctly...there's interference of some sort occurring and I wonder if anyone here can tell me what it is...

When nothing is playing on the computer (though the computer is on) there is a light/soft hum and I supposed what you could call a sort of soft crackle - for want of a better word, coming from each speaker. This increases as you turn the volume up at each speaker.

When editing in Sony Vegas and using the phones I don't hear any of the interference at all and nor did I when using the old speakers. But as soon as I plug the new speakers in I do. It happens when I'm playing the timeline in Vegas and seems to occur at the transition between frames. I have a Shuttle pro which allows me to wind down the timeline a frame at a time and I hear it when doing this...also when playing at normal speed. Once again it is more pronounced when the volume is up at the speakers.

Not sure if it has anything to do with it but rather than plugging the speakers minijack into the lead from the computer, I touch my watch with it, I get a crackle thorugh the speaker (depending on which par of the plug hits my watch). The speakers are both powered of course and when I touch my watch the power is on and the two XLR's are plugged into the speakers.

Could it be something to do with the cable or its proximity to something else?

Jay Massengill
June 8th, 2015, 04:59 AM
Computer mini-jack audio outputs are notoriously noisy with powered speakers. Plus it's probably a signal level mis-match and it's possible the cables you're using aren't correctly shielded or wired properly.

I always use a USB audio interface and good quality cables between my computer and mixer or monitor speakers.

This should cut down on noises coming from the computer and will allow you to use better connectors and cabling than a mini-jack to XLR converter cable.

If you really want 8.1, you'll need to use an interface that supports that. Or for just a low-cost stereo signal test, an interface like the Behringer UCA-222 is $30 everywhere. You would still need to convert from the RCA outputs of the Behringer to XLR if your new speaker system doesn't also have unbalanced inputs.

Rick Reineke
June 8th, 2015, 06:41 AM
The noise could have been there all the time, but was not audible on the old speakers. In addition to decent speakers, the sound card/interface is just as important.

Renton Maclachlan
June 8th, 2015, 01:44 PM
At present I don't use a sound card...son in law took it out saying the onboard sound was as good...
Board is Asus P8Z68 Delux/GEN 3...

Richard Crowley
June 8th, 2015, 01:59 PM
So do the speakers have the same background noise when the computer is turned off? Or if you simply unplug the 3.5mm plug from the computer? As Mr. Massengill mentioned, the built-in sound circuits on computers are NOTORIOUSLY noisy. It is practically impossible to make them as quiet as from your average piece of audio gear.

You might want to get that sound card back from your son in law. But even an inexpensive EXTERNAL USB sound "dongle" is typically quieter than any high-end internal sound-card. What you describe hearing is quite possibly what your son-in-law describes as "good".

Rick Reineke
June 8th, 2015, 02:06 PM
"the onboard sound was as good"
- As good as what? Consumer sound interfaces are notoriously far from flat (frequency response), and no experienced audio person would ever recommend them. Same goes for those 'designer' headphones for monitoring location audio.

Renton Maclachlan
June 8th, 2015, 02:38 PM
I have the sound card here so could put in back in again...

I also could take a recording of the sound but it would have to be tonight...I have a zoom h6 so should get a pretty good recording...

The headphones I've got are Audio-Technica ATH-M50's

Richard...When the computer is turned off there is a very faint hum but it is not the same sound...haven't checked simply unplugging the 3.5mm plug...presume it would be the same as if the computer is off as either way nothing is coming from the computer...

Not sure what my son in law means by 'good'!

Paul R Johnson
June 9th, 2015, 12:15 AM
There is a clue in the fact there's a faint noise with the computer off. If you can solder, then change the wiring on the xlr connectors to feed the ground to pin 3 and the signal to pin 2, leaving the pin 1 in connecting. I'm assuming your cable is currently wired with pin one connected to either 2 or 3 and the screen, with signal to the remaining pin? This loses the inherent balancing, but makes it susceptible to hums and noise to a small degree. Using pins 2 and 3 removes the connection to ground, and it operates in differential mode. This often helps with problem devices like PC laptops, especially Dell, who have well known interconnect problems with their power supplies. With grounds missing from such a lot of kit now, ground is rarely ground any longer.

Renton Maclachlan
June 9th, 2015, 01:51 AM
Thanks Paul...I just posted the following over at the Vegas Forum, as someone there suggested the problem was the incompatibility of mini jack (unbalanced) and XLR (balanced)...

'I thought Adam you might be onto something so rang a friend who is an audio wiz and he confirmed your analysis.

So...seeing the speakers have unbalanced RCA inputs as well, I tried linking it all up with RCA cables...
but behold, the interference is still there.

I have attached a file recorded and maximized illustrating the sound as it occurs in Sony Vegas Pro. First is normal speed down the time line, then me winding along the timeline slowly using the Shuttle Pro. The video consists of me speaking, with various graphics replacing me and being shown...some are photos, some are png files. You distinctly hear the sound change when the cursor comes to a graphic or photo...

The audio volume has been wound right back except on one small section I where wind it up a bit then down again.

Friend suggests it could be interference from the power supply, as the current varies to deal with the graphics etc and the frames down the time line...

Next step is to put the sound card back in and see if that changes anything... '

Paul R Johnson
June 10th, 2015, 12:16 AM
I posted in the other topic you started, but added stuff here. Sadly now you've posted the clip, it's clear it's bus noise, and quite common. Inside the metal box is the worst place for quiet audio. I assumed you were talking about hum, so sorry about that. My balancing trick won't work with this as the data shifting around the internal bus structure is leaking into the audio lines and it's not removable, nor is it something you are doing wrong. Some proper audio cards are properly screened, but on board chipsets aren't, nor are cheap Slot type cards. Audio performance isn't a real consideration. One card I loved started doing this when transplanted into a new computer, just one of those things. External d/a's seem immune from this because their sensitive electronics are remote from the dirty digital hash filled environment inside what is essentially a Faraday Cage. Even in mega expensive computers, the cost of the audio components is often pennies!

Richard Crowley
June 10th, 2015, 06:15 AM
Completely ordinary, common and predictable "digital hash" from inside of the computer. That is exactly why I said "the built-in sound circuits on computers are NOTORIOUSLY noisy". The only way to eliminate that is to abandon the audio outputs from the computer and use an EXTERNAL sound device. Even a $30 Behringer UCA-222 has an infinitely better chance of eliminating that noise than a $300 INTERNAL sound card. And portable "laptop" computers are even worse because of their external power supplies.

Jim Andrada
June 10th, 2015, 01:33 PM
I have an internal sound card that seems to be pretty damned noise free. On the other hand it was around $700 (Lynx L22) so I think I can agree with the comment that on-board adapters and lower cost cards are not likely to be sufficiently well shielded against PC noise.

Renton Maclachlan
June 23rd, 2015, 05:17 AM
Back again...

Interestingly...I am still getting the noise through one speaker, that by the computer (500mm from the motherboard)...but not from the one that is 1600mm from the computer....

Jon Fairhurst
June 23rd, 2015, 11:10 AM
I would definitely recommend an external, USB-based sound interface. While you can add an internal card, the industry has moved mainly to external solutions. Some nice advantages are easy access to the connectors and a nice, big volume knob within easy reach. :) Make sure to get a unit with balanced inputs and outputs. I'd get at least one XLR mic input with phantom power so you can get a mic for possible voiceovers in the future. Add some balanced lines from interface to speakers and you're set.

Renton Maclachlan
June 23rd, 2015, 01:53 PM
I have (just last week) got an external sound card...an Asus Xonar U7...

Jon Fairhurst
June 23rd, 2015, 03:06 PM
Nice. But you still get noise? Are you using balanced I/O and cables? You might try moving things around to see if there is a proximity problem.

You could also swap monitors left and right. Maybe one has a bad amp. Try swapping cables too.

If the noise is always on the same channel from the audio interface, maybe there is noise from the USB power source. Does it include an external power supply? You could try that.

If there is still a problem, the interface might be bad. Asus wouldn't have been my first choice. They're a fine motherboard company (I've bought a few), but they don't stake their reputation on having good audio. FWIW, I have a Focusrite Scarlett 2i4, which sounds great and has a great feature set. The 2i2 would be best for you as you probably don't need MIDI. The Focusrite Solo lacks balanced outputs, so I don't recommend it. The 2i4 adds MIDI, and the ability to mix the input and computer output, which is nice for multi-track recording.

Best of luck!

Paul R Johnson
June 23rd, 2015, 03:37 PM
Most of the reviews of that unit talk about how it drives headphones loudly, very few comments about audio quality. It's a gamer product, but still should be better than an internal card. How is it doing?

John Willett
June 27th, 2015, 06:16 AM
And portable "laptop" computers are even worse because of their external power supplies.

Yes - as I found recently when sorting a system for playing back evidence in a court trial - a horrible noise on the system that completely disappeared when the power supply was unplugged and the laptop was running on batteries.

Cured by changing the computer to another one.

Renton Maclachlan
June 27th, 2015, 08:16 PM
I haven't had much time to check things out recently...tried shifting the speakers around and switching cables but it didn't seem to help. When the cursor goes down the Sony Vegas timeline and hits a graphic, the interference sounds like it changes gear...

It was suggested I try a USB isolator to run my USB DAC through... HiFimeDIY USB Isolator (http://hifimediy.com/usb-isolator)

Renton Maclachlan
June 29th, 2015, 04:57 AM
Update...maybe of interest to some...

I've just been trying some things out tonight re this noise coming through my speakers….and think I may have found the cause or at least a cause...

Tonight the noise was only coming from one speaker that I could hear…and it was the same speaker regardless of which RCA lead was going to it. And it took a couple of seconds to start when playing down the Vegas time line. If I stopped the cursor and started it again immediately, the sound would start immediately. But if I stopped the cursor and waited a few seconds, it would not start immediately. That seemed odd.

However…the speakers were plugged into different power outlets…the one through which teh noise was coming was into one junction box, in close proximity to a ‘transformer/power supply’ driving speakers for my other computer, and the other into another junction box about a metre away from the other one…and which only has other plugs into it. When I plug power for each of the speakers into the latter junction box the present noise goes away.

I’m wondering if there were actually two noises…one coming from the computer power supply and one from this other power supply…or at least the junction box it was into - or something around there…[I have junction boxes galore to drive everything I have!!!]

So...maybe it is finally solved…hope so...and just after ordering a USB isolator!!! Probably won’t go amiss having the latter…

Jay Massengill
June 29th, 2015, 05:59 AM
Hard to believe none of us remembered to re-state the "rule" about all parts of the audio chain being plugged into the exact same source of power, and with all wall-wart and line-lump power supplies placed as far away as possible using additional extension cords. Of course that won't automatically solve all noise problems, but it can eliminate one pesky variable.

I hope that's the end of your troubles!

Paul R Johnson
June 29th, 2015, 07:52 AM
With our higher voltages in the UK, we've always had a slightly higher respect for mains power, but there's a creeping train of thought here that the old advice about running everything from one outlet isn't wrong - because it does often cure the problem, but worrying - because realistically, if something hums, it's down to current flowing somewhere it shouldn't. Years ago when people were stupid, people lifted the ground and hums went away. In the UK doing that now would be a major error, and if caught, trouble will follow. Grounds are for protection, and as many venue have testing requirements, or will test your kit before connecting - any cables with no ground, are going to generate a fail!

In a normal house (and I have to talk UK spec here, as I don't know the correct names for the US state of affairs) each socket is either wired direct to the consumer unit with MCBs in it, or runs in a ring out and back to another MCB. This here is the most common system. Actual distance of run is quite short, and in purely resistance terms, the ground of one socket should be as close to a short as is possible. Multiple earths are now quite rare, with most modern installations having the ground/earth tied to the neutral at the point of entry into the building, usually near the meters and consumer unit. With this kind of wiring, there really shouldn't be current flowing in the earth, and plugging into sockets in the next room should be hum and noise free. If it isn't, then something is not right, somewhere. Leakage, high resistance joints, who know?

Technical power supplies tend to be radial circuits - a common power supply fed individually to outlets, with each ground taken back to a dedicated earthing point, but with wall wart supplies, we really shouldn't be having as many ground issues as we do? All a bit odd.

Rick Reineke
June 29th, 2015, 08:35 AM
As Jay stated, it's common practice (at least in the US) to power all the gear from one circuit to minimize the possibility of ground loops. And keep power supplies at a distance and never have power cables running parallel with audio cables. Any power/audio cable crossings should be done at right-angles. (another basic audio 101)

Greg Miller
June 29th, 2015, 09:57 AM
Mr. Johnson,

This is most interesting but I'm a bit confused by your UK terminology. I would like to understand this better.

Let me share some US terminology and perhaps you can then offer a lexicon.

In a typical detached house (i.e. free-standing building, not part of a structure with multiple apartments inside) the service coming in from the utility company's pole is usually 240 VAC with a center-tapped "neutral" connection.

A metal box is mounted on the outside of the building with an integral meter socket, into which the utility company's meter is mounted.

National Electrical Code requires two metal rods (stakes) driven into the ground/earth/soil within a few feet of the meter socket. Ground wires lead from the ground rods to the meter socket, and inside the socket these actual ground wires are electrically connected (bonded) to the neutral wire coming in from the utility pole.

The two hot wires coming from the customer's side of the meter, along with the neutral/ground, are then carried to the interior of the building by a heavy cable. Inside the building, the cable goes into another metal box ("service entrance panel") which contains the main 2-pole circuit breaker, as well as the single-pole and 2-pole breakers for the various load circuits.

Inside the service entrance panel, the ground/neutral wire (from the meter socket) is connected to two separate metal buses: a ground bus, and a neutral bus. The load-carrying neutral wires that go out to various branch circuits (lights, small appliances, etc.) are white in color, and are all connected to the neutral bus. The safety ground wires that go out to the various branch circuits are green in color (or sometimes bare) and are all connected to the ground bus.

(Interestingly, our 120v circuits are all unbalanced with respect to ground, while our 240v circuits are all balanced. How does that compare with the UK?)

Perhaps you can help cross-reference some of this US terminology with UK terminology, and discuss any significant differences in practice. I am particularly puzzled by your terms like "consumer unit" and "MCBs."

--

Meanwhile, it's true that most older wall-warts should in theory minimize grounding and ground-loop issues. The exception would be those few wall-warts that do have a ground pin, in addition to the two current-carrying blades.

However, with many new wall-warts being high-frequency switchers, all kinds of strange non-sinusoidal waveforms are introduced into the power mains, and that can be a brand new source of noise ... and usually at frequencies that are much higher than mains harmonics so therefore possibly more difficult to filter out.

Paul R Johnson
June 29th, 2015, 12:25 PM
Greg - I've sort of got used to using 'ground' rather than our 'earth', and most things are actually similar, but your service entrance panel is our consumer unit. Breakers appear to be a little different. The incoming mains L+N goes to a two pole switch, and then to the two busbars. Modern installs may replace the simple switch with two separate groups, each with a 2 pole switch. Ours look like this LAP 10-Way Fully Populated Consumer Unit Dual 63A RCD | Domestic Consumer Units | Screwfix.com (http://www.screwfix.com/p/lap-10-way-fully-populated-consumer-unit-dual-63a-rcd/95181)


SKIP THIS NEXT BIT IF YOU ARE NOT INTERESTED IN UK ELECTRICS

Years ago, 45A was common as a maximum for a domestic dwelling, then it went gradually up to 100A now. MCBs - miniature circuit breakers, all to the European DIN size, and these just clip in. Our 3 pin sockets on a ring, are all 13A maximum, with fused plugs, and the ring is rated at 32A - we may have two, three or more depending on building size.

In terms of incoming cabling, virtually all private houses only have single phase electricity - so just L+N with 230V as our 'paperwork and calculation' voltage. In fact, it used to be 240V in the UK, and 220V in Europe. The EEC decided this was bad, confusing and unharmonised, so now we all use 230V. Of course, we didn't change the actual volts (mine is currently 243V, 50Hz) In the rest of europe, they still have their 220V - but Europe just fiddled the +/- allowance to make 230V the figure we all have to use. We use a star distribution system in the main - three phases, centre connected neutral, and phase to any phase is 400V (or 415 real volts!) Load is shared between the three phases as it passes down a road. Small workshops and industrial buildings get access to all three phases. We have quite a few actual input systems, but the three often quoted ones are:
TN–S
This type of electrical earthing system connects the neutral source of energy with the earth at one point only (or as near as practically possible), and with the consumer’s earthing terminal commonly connected to the metallic sheath or armour of the distributor’s service cable into the premises. With a TN–S the PE connection and N are separate conductors that are connected together only near the power source.
TN–C–S (PME)
A TN–C–S earthing system, typically known as Protective Multiple Earthing (PME), connects the supply neutral conductor of a distribution main with earth at course and at intervals along its stretch. The neutral conductor is also used to return earth faults currents safely to source by the provision of a consumer’s earthing terminal linked to the incoming neutral conductor.
TT
The neutral of the energy source is connected as with the TN–S system, however there is no provision for the consumer’s earthing, therefore they must provide their own connection to earth. This one isn't used in contemporary installs.

TN-C-S (PME) seems the most common now.

Distribution to areas is done at 11KV (or 6.6KV) using a delta system with no neutral - transformers then handle the conversion to 3 phase with neutral locally.

Domestic and small business premises don't usually have 3 phase electricity. 400V is seen as historically dangerous. It's not that many years since there were rules stating that if outlets on different phases were installed, there had to be at least 6ft between them, although that regulation thankfully went away. Clearly stuffing 400V down a cable is more of a risk, but I have dimmer packs that are three phase with multipoint connectors to the lighting circuits - and 400V between phases is maybe 8mm away!

Oddly - UK building sites require 110V tools and power for safety - cutting cables and exposing live wires. So they have to carry site transformers around.

That's a pretty quick run through!

One thing - we always do the cross at right angles recommendations, but you can now buy combined mains and XLR audio cables for powered speakers, which never seem to cause any issues! I suspect the right angle thing is perhaps now not really necessary.

Greg Miller
June 29th, 2015, 01:22 PM
Paul, thanks for that exhaustive document. I'm on my way to work now but I'll try to digest it slowly and thoroughly later tonight.

A few quick questions:

You say "L+N goes to a 2 pole switch."

1.) In USA, "N" means neutral and is at ground/earth potential. Is that the same in UK?

2.) Does a typical residential house have just one "hot" line coming in, or two?

3.) *IF* your Neutral is at ground potential, does it still go through a switch in the "consumer unit" (entrance panel)? In USA we never switch or disconnect Neutral; in fact I believe it's illegal to do so.

I could go on and on, each question leads to another. It seems that "standard practice" is very different here from there, but maybe some of that will resolve when I come to a better understanding of all the details.

By the way, the 3-wire residential service I described above is called "split phase" because it is from just one phase of the high voltage distribution system. But it's center tapped so the 240 volt secondary is "split" in half. (The two hot legs are 180º out of phase with each other.)

In a bigger apartment building or commercial building, it's usual to bring in all three phases. Most newer construction is "Y" connected with the center of the Y being neutral/earth/ground. Any one hot leg is 120 volts above neutral. So if most of the loads are 120 volt, it's still possible to try to balance the phases. The voltage between any two hot legs is 208 volts (since they are 120º apart in phase).

Some older commercial buildings had a "delta" connection. Two of the transformer secondaries were simply 240 volts. The third secondary was 240 volt center tapped. That center tap was connected to neutral/ground. This allowed a true 240 volts on all three phases, to run higher-power loads. But only two of the legs could provide 120 volts to neutral. The third leg was high with respect to neutral and so could be used only with respect to the other two hot legs. Delta is always a PITA because of the potential danger with that high leg. Do you have such a thing? Was that explanation clear?

Paul R Johnson
July 4th, 2015, 09:53 AM
I appreciate we've kind of hijacked the topic, but as tricks and tips for reducing electrical noise (and of course shocks) are handy and common, these differences might be useful - perhaps split out somewhere else?

At the point the supply enters the home, in most cases, the N- Neutral is bonded to the ground - which is then distributed from this point. The Live and Neutral BOTH go through the two pole switch, so with it off - there is ground at all outlets, but both L+N are isolated. RCDs are now quite common to give extra protection, downstream of the 2 pole switch.

Phase wise, we only have 120 degree phase shift. Occasionally, two phases may be in one building - if perhaps a domestic home is extended, and additional power is required, needing an extra feed from the supply in the street. Planned higher loads will be always 3 phase, with three phases and Neutral being brought in. Connector wise - we use similar looking ones for industrial applications CEE17 will bring up the common ones in Google - Colour coded for voltage - yellow 110V, Blue 230V, Red 400V - most versions of yellow and blue are 3 pin, single phase. Red has 5 pins - E (ground)-L1-L2-L3-N, although some motors will use a 4 pin Delta wired Red connector. There's even a similar purple one for less than 50V usage.

People who use video in the events world often have domestic 13A fused plug adaptors to 16A CEE17 outlets - as these are common in studios, theatres and other pro venues. (to confuse things, they come in 16A/32A/63A and 125A versions too!)

I believe there is a very similar looking connector used in the US as some of the lighting kit I have imported from China has a Blue connector fitted - but it's not our spec in size and pin spacing.

Quite a few of my colleagues have taken kit to the US and been able to find 230ish volts - and I could never work out how - Using the UK 3 phase system, on US voltages, I couldn't find a phase to phase voltage anywhere near 230 - so your split phase system, 180 degrees apart is totally new to me, and of course, the answer! Here, phase to phase connections are rare unless you are in big venues, but much more common in the US.

As another bit of history - our protection devices that monitored leakage to earth used to be a three terminal device, that measured how much was leaking to ground and used that to trip the circuit as the leakage must be a fault. Now, we normally use two terminal devices that monitor the difference between L and N, again, on the assumption that if they are not the same, it's going somewhere else - which is bad!

Richard Crowley
July 4th, 2015, 10:37 AM
Its not all that mysterious. In North America, center-tapped, single-phase 230V is delivered to domestic customers and the center-tap is locally grounded and called "neutral". So typically half the 115V circuits in the house come from one side ("phase") of the transformer secondary winding, and the other half of the circuits are connected to the other side.

High-power devices (stoves/"cookers"), water heaters, etc. are designed to use 230V which is taken across the two "phases". I agree that use of the term "phase" for a simple center-tapped source seems rather pretentious. But that is what electricians have been calling it for several generations.

3 phase power is almost never found in domestic premises (homes, apartments, etc.) But it is commonly provided to larger industrial customers.

Rick Reineke
July 4th, 2015, 10:38 AM
"Quite a few of my colleagues have taken kit to the US and been able to find 230ish volts - and I could never work out how"
- They likely tied directly into the main electrical breaker box which typically enters @ 220 volts / three phase, then broken out to multiple 110 volt two phase circuits that feed the US "Edison" wall sockets. Most lighting techs here as well tie directly into the three phase mains.

Paul R Johnson
July 4th, 2015, 02:02 PM
It just never occurred to me that there was any other system than 3 phase - the notion of a centre tapped transformer just never occurred to me for mains power.

Richard Crowley
July 4th, 2015, 09:57 PM
It just never occurred to me that there was any other system than 3 phase - the notion of a centre tapped transformer just never occurred to me for mains power.
There are actually power schemes with MORE than 3 phases (particularly very high power circuits, and works where they convert between AC and DC (and back again), etc. But 3-phase (120 degree) power is virtually unknown outside industrial sites in North America.

Greg Miller
July 5th, 2015, 09:04 PM
In North America, center-tapped, single-phase 230V is delivered to domestic customers and the center-tap is locally grounded and called "neutral". So typically half the 115V circuits in the house come from one side ("phase") of the transformer secondary winding, and the other half of the circuits are connected to the other side. {snip}
I agree that use of the term "phase" for a simple center-tapped source seems rather pretentious. But that is what electricians have been calling it for several generations.

3 phase power is almost never found in domestic premises (homes, apartments, etc.) But it is commonly provided to larger industrial customers.

Where I've worked, I don't recall hearing anyone referring to to the two 120 volt hot lines as "phases." I've always heard the term "leg." So there are two "legs" of one phase, the two legs being 180º apart.

Of course in an older delta system, where one transformer secondary is center tapped, the two primary terminals of that transformer are connected to two phases of the high-voltage distribution system. But the voltage across the secondary of that transformer is still just one phase and those two 120 volt legs are still 180º apart. So at first glance this might seem to be a discrepancy ... but it really isn't. The primary of that step-down transformer is connected to two [primary] phases, but the secondary voltage is just a single phase (center-tapped) sine wave.

BTW, around here most of the multi-story multi-unit apartment buildings are supplied with 3-phase 120v-208v Y service. In that case, each apartment unit (i.e. single residence unit) may be supplied with only two of the phases (unless there is a high-current 3-phase load, such as a large air-conditioning compressor). If you compare any two phases with respect to neutral you'll find that they're 120º apart, each being 120 volts above neutral. But if you connect a load between any two phases, you will see a single-phase sine wave between the legs with a voltage difference of 208 volts.

Greg Miller
July 5th, 2015, 09:19 PM
But 3-phase (120 degree) power is virtually unknown outside industrial sites in North America.

Perhaps that depends on your definition of "industrial sites." {my dictionary says "industrial: of or pertaining to, or resulting from, industry or productive labor"}

As I stated above, a lot of apartment buildings in this area (typically a few hundred separate apartments in the building) do have three phase coming in. However, any given apartment unit (i.e. occupied by one person or family) might have only two of the three legs, since they have no 3-phase loads.

But many small [non-industrial] businesses have 3-phase service. e.g. movie theatres, coin laundries (laundromats), restaurants, retail stores, supermarkets, banks, etc. I don't think of any of these as being an "industrial site" but I know from my own experience that they do have 3-phase service.

There is one very old business corridor in this town (located on either side of a long alley) where the service is older 3-phase delta. Most of the businesses on that corridor (except for the very small shop fronts) have all three phases coming in, because they have some significant number of 240-volt loads. But of course there's always a problem with load balancing because the 120-volt loads can be connected to only one of the three step-down transformers (i.e. between two of the three high-voltage primary phases).

Most of the newer businesses here have 3-phase Y, with all three legs coming into the premises. The Y connection makes load balancing much easier (since each of the three legs can supply 120v to neutral). The only concession is that higher power loads need to run on 208 instead of 240; however 208 volt equipment has been almost ubiquitous for the last 30 years or more.

Richard Crowley
July 5th, 2015, 11:47 PM
I was imprecise in my language. By "industrial" I was trying to convey the concept of non-domestic.

Certainly most power in North America is DISTRIBUTED as three 120-degree phases. But, it is DELIVERED to domestic end-users as 240V center-tapped "single-phase". Indeed, apartment buildings with more than a few dozen units probably have one or more large transformers that take the 440V (or whatever) 3-phase power and step it down to many 240V feeds which have their center-taps grounded and delivered to the individual unit meters and circuit breaker boxes Virtually no consumer electrical appliances are designed handle the voltages produced by 120 degree phases (208V or 277V)

Paul R Johnson
July 6th, 2015, 12:32 AM
With two different voltages available, presumably from different socket types, is there a constant stream of appliances being damaged? I see American equipment with IEC type connectors, are there rules to stop idiots or the careless blowing things up? People here just see a cable laying in the floor and plug it in. I'd hate to think what we would do with dual voltages in homes!

I'd never realised the differences between us were as great as this.

Greg Miller
July 6th, 2015, 06:28 AM
Almost all residential/consumer equipment operates on 120v, and 240v connectors are significantly different, so it's almost impossible to make the kind of mistake that you envision.

There are two types of connectors in general use for 120v. The 5-15 socket is for use up to 15 amps, the 5-20 socket is for use up to 20 amps. You can find some photos here:
https://en.wikipedia.org/wiki/NEMA_connector .
Note that the 5-20 socket has one T-shaped terminal.

Most residential/consumer equipment draws 15 amps or less, and uses a 5-15 plug with two parallel blades; this plus will fit either a 5-15 or 5-20 socket.

Some slightly larger 120v equipment (e.g. a commercial floor polisher or larger room-size air conditioner) uses a 5-20 plug, with the two current-carrying blades at 90º to each other; this plug will fit only a 5-20 socket.

There is also the question of differentiating between the line (hot) and neutral side of the circuit. The neutral blade is the same thickness as the "hot" blade, but the neutral blade is wider. Such connectors are called "polarized." A polarized plug can be inserted into a polarized receptacle only in the correct orientation. This distinction is required because NEC prohibits switching the neutral connection; only the hot connection may be switched. Thus it's important that the unswitched side of the device is always connected to the neutral side of the circuit.

National Electrical Code permits any given branch circuit (i.e. a circuit that is protected by a given circuit breaker) to carry a continuous load of no more than 80% of the rated size of the circuit breaker. Thus a 20A circuit can legally supply a continuous load of only 16 amps. With nominal voltage of 120v, that's a continuous wattage of 1920 watts.

NEMA 5 connectors are never used for 240v. When you go to the higher voltage, you go to NEMA 6 or larger. There is little chance of confusing anything else with NEMA 5, and it is impossible to insert a 120v NEMA 5 plug into any type of 240 volt socket.

With few exceptions, the only residential/consumer appliances that require the higher voltage are electric clothes dryers and electric cooking ranges. These use NEMA 14 connectors which can't be mistaken for anything else.

Electric water heaters almost always use 240 or 208, but they are hard-wired and do not have any plug/socket connections, so no chance of an error there.

Paul R Johnson
July 6th, 2015, 11:05 AM
What is the most c common type of male connector on appliances - here, we have on domestic kit 3 pin connectors for most connections - with the equipment end the figure of 8 shape that can be reversed so L+N can be connected the wrong way around. This connection has the ground pin of the connector plug usually made from plastic - which is necessary to open up the shutters all our outlets have in homes. This kind of equipment does not have a ground connection, and must not be fitted with one - the case is isolated from the power supply. Three pin connections are done via IEC connectors.
http://static.rapidonline.com/catalogueimages/Module/M063584P01WL.jpg
http://uk.farnell.com/productimages/large/en_GB/42409699.jpg

More and more professional gear is now being fitted with Powercon type locking connectors
http://uk.farnell.com/productimages/large/en_GB/AV25118-40.jpg

Richard Crowley
July 7th, 2015, 12:33 PM
What is the most c common type of male connector on appliances?

Virtually everything that draws 15A or less uses one of these mains plugs.
The 2-pin variety are still widely used in "double-insulated" gadgets